MyRocks: How Facebook Cut MySQL Storage Nearly in Half
Why Facebook replaced InnoDB with MyRocks, an LSM-tree storage engine built on RocksDB, to shrink storage footprint across its huge MySQL fleet.
Engineering at scale
12 articles · 48 min of reading
Why Facebook replaced InnoDB with MyRocks, an LSM-tree storage engine built on RocksDB, to shrink storage footprint across its huge MySQL fleet.
How Facebook's Ads team, wrestling with cascading manual DOM updates, built React around declarative rendering and a virtual DOM diff.
How Facebook built Scuba, an in-memory analytics engine that trades storage cost and precision for sub-second answers engineers need while debugging live incidents.
How Facebook forked Google's LevelDB into RocksDB to target fast flash storage, and how an embedded engine built for one problem became infrastructure for many.
How Facebook's 2012 mobile rewrite exposed REST's limits and led to GraphQL, a query language built around what the client actually needs.
How Facebook built Presto to give analysts interactive, seconds-scale SQL queries over a Hive warehouse that Hive's own engine took minutes to hours to query.
How Facebook's Thrift let services written in different languages call each other efficiently, years before gRPC popularized the same idea.
How Facebook's Haystack replaced a filesystem-per-photo approach with a purpose-built object store to eliminate the metadata bottleneck of billions of images.
Why generic memcache stopped fitting Facebook's social graph, and how TAO's objects-and-associations model became the read path for billions of requests.
How Facebook rebuilt PHP's execution model with HHVM and added a type system with Hack, without forcing a rewrite of its enormous existing codebase.
How Facebook turned a simple key-value cache into a cluster of thousands of memcached servers without drowning in stale reads and thundering herds.
How Facebook built Cassandra to solve inbox search at scale, then open sourced a database that outgrew the problem it was designed for.